Pith. sign in

REVIEW 2 major objections 4 minor 85 references

The paper claims a protocol that keeps every control pulse finite and singularity-free for arbitrary qubit targets, and that the noise-optimal member of the resulting pulse family prepares high-fidelity states under non-Markovian noise.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

T0 review · deepseek-v4-flash

2026-08-04 09:25 UTC pith:EKTVHXD6

load-bearing objection Useful trajectory-splitting construction for bounded invariant-based pulses, but the singularity-free proof is not airtight and the noise results are in-sample. the 2 major comments →

arxiv 2510.15340 v3 pith:EKTVHXD6 submitted 2025-10-17 quant-ph cs.LGcs.SYeess.SY

Singularity-free dynamical invariants-based quantum control

Ritik Sareen , Akram Youssry , Alberto Peruzzo This is my paper
classification quant-ph cs.LGcs.SYeess.SY
keywords quantum state preparationdynamical invariantsshortcuts to adiabaticitysingularity-free controlnon-Markovian noiseinverse engineeringmachine-learning quantum controlrandom telegraph noise
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper tries to establish that the singularity problem that has dogged invariant-based quantum control can be removed entirely. By splitting the trajectory between initial and target states into subtrajectories and switching the reference axis at each stage, the authors keep the invariant's denominator component strictly away from zero, so every control pulse their method produces is bounded in amplitude, for any target qubit state, including antipodal ones that previously forced pulses to diverge. The second claim is that the resulting family of equally correct closed-system pulses contains members that stay accurate under colored, non-Markovian noise, the regime where Lindblad master equations fail. For known noise a perturbative cost function finds the best member; for unknown noise a grey-box machine-learning model, trained on nothing but pulse parameters and measured expectation values, does the same. In simulations with strong random-telegraph noise, the protocol prepares six diverse targets at roughly 92-98% fidelity, and the noise-optimized pulses beat the best of 10,000 random pulses in nearly every case. A sympathetic reader would care because this turns invariant-based shortcuts to adiabaticity from a closed-system curiosity into a practical route for state preparation on noisy real-world hardware.

Core claim

The central claim is Theorem 1: every control pulse the method produces is bounded, finite in amplitude for all t in [0,T]. The argument reduces boundedness to keeping the invariant's reference component f3(t) real and non-vanishing. Trajectory splitting chooses a reference axis per subtrajectory so the reference Hamiltonian coefficient has the same sign at both endpoints, forcing f3 to keep one sign throughout, while the numerically computed bound vmax on the free polynomial parameters keeps f3 strictly positive in between; h1 = (f1h3 - df2/dt)/f3 and h2 = (f2h3 + df1/dt)/f3 are then finite and continuous. From the resulting family, a noise-optimal pulse is selected, via a perturbative cost

What carries the argument

The central object is the Lewis-Riesenfeld dynamical invariant for a qubit, I(t) = (f1(t)sigma1 + f2(t)sigma2 + f3(t)sigma3)/2, a Hermitian operator whose eigenstates evolve exactly under the control Hamiltonian and whose coefficient vector has conserved magnitude f1^2 + f2^2 + f3^2 = c^2. Given a reference pulse h3(t), the remaining controls are h1 = (f1h3 - df2/dt)/f3 and h2 = (f2h3 + df1/dt)/f3, so all singularities reduce to one condition: f3 must stay nonzero. Trajectory splitting enforces it by switching the reference axis so f3 has one sign at both ends of each subtrajectory; the bound vmax on the free parameters of the degree-n polynomial ansatz for (f1,f2) keeps f3 from vanishing in

Load-bearing premise

The singularity-free guarantee rests on the numerically computed bound vmax: the paper evaluates a per-time quadratic inequality on a discrete grid of time points and assumes its minimum is a valid bound at every continuous time in between, so if f3(t) were to cross zero between grid points, the promised bounded pulses would diverge.

What would settle it

Take a target from case (vi) (population inversion), set the free parameters to (1-epsilon)vmax, and evaluate the invariant and pulse amplitudes on a grid 100 times finer than the one used to compute vmax: any sample with f3(t) <= 0, or with |h1(t)| or |h2(t)| exceeding a preset finite cap, refutes Theorem 1. Experimentally, implement the grey-box-optimal pulse for that target on a qubit whose environment is engineered to produce random-telegraph noise and compare the measured final fidelity with the predicted roughly 92-96%.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • Any single-qubit target state, including antipodal states such as population inversion that previously forced invariant-based pulses to diverge, can be prepared with bounded, continuous, low-bandwidth control fields.
  • Invariant-based control now covers non-Markovian open systems: the optimal-pulse selection works for noise that admits no Lindblad master equation, such as multi-axis colored random-telegraph noise.
  • The method returns an infinite family of equally correct closed-system pulses, so the same framework can be repurposed for other objectives, minimum energy, bandwidth limits, or leakage avoidance, simply by changing the cost function.
  • In the simulations the optimized pulses outperform the best member of a 10,000-pulse random family for all six targets in the known-noise case and for nearly all in the unknown-noise case, reaching fidelities of roughly 92-98% under strong colored noise.
  • The grey-box machine-learning route needs only experimentally accessible data, pulse parameters and Pauli expectation values, so no master equation or noise model is required to use it.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Because vmax is found by minimizing the bound of a quadratic inequality over a discrete time grid, Theorem 1's every-t guarantee inherits a discretization gap: an analytic or interval-certified bound, or a grid-refinement check at the extreme parameters (1-epsilon)vmax, would close the only gap between the heuristic and a fully rigorous guarantee.
  • The trajectory-splitting logic should transfer wherever an invariant lives on a sphere: qudits and multi-qubit systems face the same denominator-vanishing singularities, and the reference-axis switching idea appears dimension-agnostic; the authors list such extensions as future work.
  • The grey-box loop could run directly on real hardware as a self-calibrating protocol, since its training data are just pulse parameters and measured Pauli expectations; the same optimization could be executed without any simulation or noise model.
  • The purity analysis suggests a testable refinement: adding a purity term to the cost function might recover fidelity where a mixed state and a pure state share the same fidelity with the target, a possibility the authors themselves note; one could check whether the known-noise versus unknown-noise gap for the antipodal target narrows with such a term.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

2 major / 4 minor

Summary. The paper proposes an invariant-based inverse-engineering protocol for qubit state preparation in the presence of non-Markovian noise. The protocol splits the Bloch-sphere trajectory into subtrajectories, parameterizes the dynamical invariant by high-degree polynomials with bounded free variables, and selects the control pulse that best mitigates noise using either a whitebox cost built from a second-order Dyson expansion or a graybox machine-learning model. Numerical simulations are reported for six target states under two-axis random telegraph noise, with the claim that all produced pulses are guaranteed to be bounded (singularity-free) and that high-fidelity state preparation is achieved under multi-axis classical colored noise.

Significance. If the two main claims can be made rigorous, the paper would be a useful contribution to shortcut-to-adiabaticity and quantum control: the trajectory-splitting construction is a sensible way to avoid singular denominators in invariant-based pulse design, and the combination with a graybox noise-mitigation layer is practically relevant. The paper also ships a clean analytic parameterization and a transparent optimization workflow. However, the headline boundedness theorem currently rests on a discretized numerical certificate, and the reported noise-mitigation fidelities are computed with the same approximate Dyson model used for both the whitebox cost and the graybox training data. These gaps are load-bearing for the paper's central claims.

major comments (2)
  1. [Sec. IV D, Eq. (56), and Theorem 1] Theorem 1's assertion that all pulses are bounded hinges on the bound vmax computed by discretizing s and taking the minimum of the positive roots V2(s*) of Eq. (56). This procedure certifies f3(s)>0 only at the sampled grid points. f1 and f2 are degree-18 polynomials, so f1^2+f2^2 can dip below c^2 between grid points; the (1−epsilon) rescaling protects only the sampled points. The proof of Theorem 1 states that Section IV D 'ensures' f3(t) is real and non-vanishing, but Section IV D itself describes a heuristic numerical scheme. Without an analytic uniform bound, a Lipschitz estimate, or a grid-refinement certificate, the continuous-time singularity-free guarantee is not established. If the radicand in Eq. (33) becomes negative between grid points, f3 becomes complex or zero and Eq. (26) gives nonphysical or unbounded pulses. The values in Supplementary Table I inherit this limitation.
  2. [Sec. IV F, Sec. V (Dataset generation), Table III] The noise-mitigation results are evaluated with the same second-order Dyson expansion used to construct the whitebox cost in Eq. (62) and to generate the graybox training data. Thus the fidelities reported in Table III are not an independent test of the protocol against the actual RTN dynamics; the optimizer is scored against the same approximate model it is optimizing. The Discussion acknowledges that the whitebox model is 'perfectly matched to the simulation model,' but the graybox model inherits the same circularity because its training labels come from the same Dyson simulation. To support the claim of high-fidelity state preparation under multi-axis classical colored noise, the authors should benchmark against an independent exact or higher-order method (for example, stochastic Schrodinger dynamics over RTN realizations) and report convergence of the Dyson truncation, especially at
minor comments (4)
  1. [Abstract and Section II] The abstract first says 'finite-dimensional state preparation' and later restricts to 'single-qubit state preparation'; please clarify the exact scope in the introduction as well.
  2. [Section IV D] The phrase 'liner combination' should read 'linear combination'. Also, Eq. (44) is written in t while the subsequent derivation uses normalized time s; please make the switch explicit to avoid ambiguity.
  3. [Section V, Implementation] The number of grid points used to compute vmax is not stated. Because the boundedness claim depends on this discretization, please report the grid size and show that the resulting vmax is stable under refinement.
  4. [Discussion] The claim that the method is 'mathematically guaranteed to yield optimal solutions' is stronger than what is demonstrated, since the noise mitigation step uses a random search over an approximate cost function. Please temper this statement or specify the precise sense of optimality.

Circularity Check

2 steps flagged

Open-system 'high fidelity' results reduce to the same second-order Dyson model used to define the cost; the graybox is trained and benchmarked on that same surrogate. The closed-system bounded-pulse construction is independent.

specific steps
  1. fitted input called prediction [Sec. IV F (Eqs. 59, 62); Sec. V 'Dataset generation'; Supp. Notes 2 and 6]
    "In this paper, we use Dyson expansion up to the second order to estimate the noise operator [62], as defined in Supplementary Note 2. ... In this paper, we use Dyson expansions truncated to second-order in the coupling strength for simulating the system dynamics (See Supplementary Note 6 for detailed computations)."

    Eq. (62) defines J(T;Theta) as 1 - F(T) for the final state computed from the same second-order Dyson expansion used for the 'simulation' benchmark. The optimal Theta is chosen by minimizing this J, and Table III then reports the fidelity of that optimal pulse. Thus the reported high fidelity is simply the minimized objective read back from the same model, not an independent evaluation of the pulse under the actual colored-noise dynamics.

  2. fitted input called prediction [Sec. V 'Dataset generation'; Sec. IV F 'graybox'; Sec. VI Discussion]
    "The dataset serves two purposes. The first is to provide a benchmark for the performance of different pulses. This is done by computing the fidelity for each pulse using Equation 3, and then plotting the histogram of those fidelities. The second purpose is to train the graybox model for the noise mitigation step. ... the whitebox model used here is perfectly matched to the simulation model, a condition rarely met in practical settings."

    The graybox is trained on labels generated by the same Dyson-truncated simulator that later serves as the benchmark for Table III. Its 'optimal' pulse is then evaluated with that same simulator. Consequently, the claimed unknown-noise robustness is validated only against the surrogate model that generated the training data, providing no independent check against exact dynamics or experiment; the paper's own caveat about the whitebox/simulation match confirms this.

full rationale

The closed-system inverse-engineering chain is self-contained and not circular: Eq. (26) follows algebraically from the invariant condition (14), the boundary conditions fix the polynomial coefficients, and the trajectory-splitting construction is an independent design step. The skeptical concern about Theorem 1 (discretized vmax in Section IV D) is a correctness gap, not a circularity: sampling s isin [0,1] and taking min of V2(s*) does not by itself prove f3(t) > 0 at unsampled times, so the continuous-time guarantee is under-supported, but this is an omitted certificate rather than a definitional reduction. The circularity is in the open-system demonstration: Eq. (62)'s cost is exactly 1 minus the fidelity computed from the same second-order Dyson expansion used for simulation, and the graybox is trained on and benchmarked against that same Dyson surrogate. Thus the reported high-fidelity 'predictions' are, by construction, the optimized cost values rather than independently verified outcomes. Score 6 rather than 8 or 10 because the central bounded-pulse parameterization and the exact closed-system state-preparation result retain independent mathematical content; the noise-stage reduction is partial circularity.

Axiom & Free-Parameter Ledger

3 free parameters · 4 axioms · 0 invented entities

The central closed-system construction is self-contained apart from standard invariant theory and the full-control/|0⟩ assumptions. The main external postulate is the Dyson-truncated noise model, which is used both to optimize and to evaluate the noise-mitigation results, so it is not independently benchmarked.

free parameters (3)
  • vmax bound on invariant polynomial free variables = 0.1655c to 0.5739c depending on subtrajectory (Supplementary Table 1)
    Computed numerically by discretizing s ∈ [0,1] and taking the minimum of the positive root of Eq. (56). This bound is load-bearing for Theorem 1, but no analytic uniform bound is proven.
  • Polynomial degree n = 18
    Chosen ad hoc in Section V to give 30 free parameters per subtrajectory; higher degree is claimed to provide more pulse-shape flexibility.
  • epsilon margin = 1e-5
    Used in Section IV D to rescale vmax so f3^2 stays strictly positive in numerics.
axioms (4)
  • standard math Lewis-Riesenfeld invariant condition (Eq. 4) and the boundary condition [I(tb), H(tb)] = 0 (Eq. 12) guarantee ground-state-to-ground-state transfer.
    This is the standard invariant-based inverse-engineering framework cited from Refs. [28-32].
  • domain assumption Full three-axis control of the qubit is available.
    Stated in Section II as necessary for bounded pulses on all target states; not all experimental platforms provide full x-y-z control.
  • domain assumption Initial state is always |0⟩, the ground state of Hctrl(0) = -Ωσz.
    The trajectory-splitting cases in Section IV A are derived for this specific initial state; the abstract's 'arbitrary initial state' is not demonstrated.
  • ad hoc to paper The second-order Dyson expansion of the noise operator V_O accurately describes the RTN noise dynamics at gx/Γ = 10 and gz/Γ = 13.
    Used in both the whitebox cost and the dataset generation (Supp. Note 6); no exact simulation is provided to validate this strong-coupling approximation.

pith-pipeline@v1.3.0-alltime-deepseek · 33702 in / 12781 out tokens · 107494 ms · 2026-08-04T09:25:01.118270+00:00 · methodology

0 comments
read the original abstract

State preparation is a cornerstone of quantum technologies, underpinning applications in computation, communication, and sensing. Its importance becomes even more pronounced in non-Markovian open quantum systems, where environmental memory and model uncertainties pose significant challenges to achieving high-fidelity control. Invariant-based inverse engineering provides a principled framework for synthesizing analytic control fields, yet existing parameterizations often lead to experimentally infeasible, singular pulses and are limited to simplified noise models such as those of Lindblad form. Here, we introduce a generalized invariant-based protocol for finite-dimensional state preparation under arbitrary noise conditions. We transform the finite-dimensional control problem into the equivalent problem for a single-qubit, by restricting the dynamics to a designed SU(2) subspace. The control protocol then proceeds in two-stages: first, we construct a family of bounded pulses that achieve perfect state preparation in a closed system; second, we identify the optimal member of this family that minimizes the effect of noise. The framework accommodates both (i) characterized noise, enabling noise-aware control synthesis, and (ii) uncharacterized noise, where a noise-agnostic variant preserves robustness without requiring a master-equation description. Numerical simulations demonstrate high-fidelity state preparation across diverse targets while producing smooth, hardware-feasible control fields. This singularity-free framework extends invariant-based control to realistic open-system regimes, providing a versatile route toward robust quantum state engineering on NISQ hardware and other platforms exhibiting non-Markovian dynamics.

Figures

Figures reproduced from arXiv: 2510.15340 by Akram Youssry, Alberto Peruzzo, Ritik Sareen.

Figure 1
Figure 1. Figure 1: FIG. 1 [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2 [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3 [PITH_FULL_IMAGE:figures/full_fig_p012_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4 [PITH_FULL_IMAGE:figures/full_fig_p013_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: FIG. 5 [PITH_FULL_IMAGE:figures/full_fig_p014_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: FIG. 6 [PITH_FULL_IMAGE:figures/full_fig_p015_6.png] view at source ↗
Figure 1
Figure 1. Figure 1: FIG. 1 [PITH_FULL_IMAGE:figures/full_fig_p022_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2 [PITH_FULL_IMAGE:figures/full_fig_p025_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3 [PITH_FULL_IMAGE:figures/full_fig_p026_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4 [PITH_FULL_IMAGE:figures/full_fig_p026_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: FIG. 5 [PITH_FULL_IMAGE:figures/full_fig_p027_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: FIG. 6 [PITH_FULL_IMAGE:figures/full_fig_p027_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: FIG. 7 [PITH_FULL_IMAGE:figures/full_fig_p028_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: FIG. 8 [PITH_FULL_IMAGE:figures/full_fig_p029_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: FIG. 9 [PITH_FULL_IMAGE:figures/full_fig_p030_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: FIG. 10 [PITH_FULL_IMAGE:figures/full_fig_p031_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: FIG. 11 [PITH_FULL_IMAGE:figures/full_fig_p032_11.png] view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

85 extracted references · 3 linked inside Pith

  1. [1]

    Initial state to intermediate state with Z as refer- ence axis, with hz(t) = hz(0), ∀t ∈ [0,T/ 2]

  2. [2]

    • Case 4: The target state is antipodal with re- spect to initial state, or hz(0)hz(T ) ≤ 0, hx(T ) = 0 and hy(T ) = 0

    Intermediate state to target state with Y as refer- ence axis, with hy(t) = hy(T ), ∀t ∈ [T/ 2,T ]. • Case 4: The target state is antipodal with re- spect to initial state, or hz(0)hz(T ) ≤ 0, hx(T ) = 0 and hy(T ) = 0. Here we need to construct three subtrajectories between [0 ,T/ 3], [T/ 3, 2T/ 3], and [2T/ 3,T ]. The intermediate Hamiltonians are chose...

  3. [3]

    Initial state to first intermediate state with Z as reference axis and hz(t) = hz(0), ∀t ∈ [0,T/ 3]

  4. [4]

    First Intermediate state to second intermediate state with X as reference axis and hx(t) = hz(0), ∀t ∈ [T/ 3, 2T/ 3]

  5. [5]

    whitebox

    Second intermediate state to target state with Z as reference axis and hz(t) = hz(T ), ∀t ∈ [2T/ 3,T ]. These four cases cover all possible target states for a single qubit, starting from the initial state |0⟩. Similar logic can be followed to design the subtrajectories given a different initial state. The outcome of this step is finding the appropriate bou...

  6. [6]

    Advances in quantum metrology

    Vittorio Giovannetti, Seth Lloyd, and Lorenzo Mac- cone. Advances in quantum metrology. Nature photonics, 5(4):222–229, 2011

    2011

  7. [7]

    and its variants, a widely adopted set of algorithms, TABLE III: Fidelity(%) of optimal control pulses. We compare the fidelity (expressed as percentage) of the optimal control pulses obtained through whitebox (WB) and graybox (GB) approaches against the minimum, average, and maximum fidelity observed in the dataset, for each target. ID Min Avg Max WB GB (i...

  8. [8]

    Quantum computing

    Andrew Steane. Quantum computing. Reports on Progress in Physics , 61(2):117, 1998

    1998

  9. [9]

    Quantum algorithms: an overview

    Ashley Montanaro. Quantum algorithms: an overview. npj Quantum Information , 2(1):15023, 2016

    2016

  10. [10]

    Quantum communication

    Nicolas Gisin and Rob Thew. Quantum communication. Nature photonics, 1(3):165–171, 2007

    2007

  11. [11]

    Quantum error cor- rection

    Daniel A Lidar and Todd A Brun. Quantum error cor- rection. Cambridge university press, 2013

    2013

  12. [12]

    Quantum sensing

    Christian L Degen, Friedemann Reinhard, and Paola Cappellaro. Quantum sensing. Reviews of modern physics , 89(3):035002, 2017

    2017

  13. [13]

    Raimond, and Serge Haroche

    Clment Sayrin, Igor Dotsenko, Xingxing Zhou, Bruno Peaudecerf, Tho Rybarczyk, Sbastien Gleyzes, Pierre Rouchon, Mazyar Mirrahimi, Hadis Amini, Michel Brune, J. Raimond, and Serge Haroche. Real-time quantum feedback prepares and stabilizes photon number states. Nature, 477:73–7, 2011

    2011

  14. [14]

    Schulte- Herbrueggen, and Steffen Glaser

    Navin Khaneja, Timo Reiss, Cindie Kehlet, T. Schulte- Herbrueggen, and Steffen Glaser. Optimal control of coupled spin dynamics: Design of nmr pulse sequences by gradient ascent algorithms. Journal of magnetic resonance (San Diego, Calif. : 1997) , 172:296–305, 03 2005

    1997

  15. [15]

    Chopped random-basis quantum optimization

    Tommaso Caneva, Tommaso Calarco, and Simone Mon- tangero. Chopped random-basis quantum optimization. Phys. Rev. A , 84:022326, Aug 2011

    2011

  16. [16]

    One decade of quantum optimal control in the chopped random basis

    Matthias M Mller, Ressa S Said, Fedor Jelezko, Tommaso Calarco, and Simone Montangero. One decade of quantum optimal control in the chopped random basis. Reports on Progress in Physics , 85(7):076001, 2022

    2022

  17. [17]

    A variational eigenvalue solver on a quantum processor

    Alberto Peruzzo, Jarrod Mcclean, Peter Shadbolt, Man Hong Yung, Xiaoqi Zhou, Peter Love, Aln Aspuru- Guzik, and Jeremy O’Brien. A variational eigenvalue solver on a quantum processor. Nature communications, 5, 2013

    2013

  18. [18]

    A quantum approximate optimization algorithm

    Edward Farhi, Jeffrey Goldstone, and Sam Gutmann. A quantum approximate optimization algorithm. 2014

    2014

  19. [19]

    Modelling and feedback control design for quantum state preparation

    Ramon van Handel, John K Stockton, and Hideo Mabuchi. Modelling and feedback control design for quantum state preparation. Journal of Optics B: Quantum and Semi- classical Optics, 7(10):S179, 2005

    2005

  20. [20]

    J. J. Garc´ ıa-Ripoll and J. I. Cirac. Quantum computation with unknown parameters. Phys. Rev. Lett. , 90:127902, 2003

    2003

  21. [21]

    V. V. Sivak, A. Eickbusch, H. Liu, B. Royer, I. Tsioutsios, and M. H. Devoret. Model-free quantum control with reinforcement learning. Phys. Rev. X , 12:011059, 2022

    2022

  22. [22]

    Clas- 17 sifying global state preparation via deep reinforcement learning

    Tobias Haug, Wai-Keong Mok, Jia-Bin You, Wenzu Zhang, Ching Eng Png, and Leong-Chuan Kwek. Clas- 17 sifying global state preparation via deep reinforcement learning. Machine Learning: Science and Technology , 2(1):01LT02, 2020

    2020

  23. [23]

    Deep Reinforcement Learning for Quantum State Preparation with Weak Nonlinear Mea- surements

    Riccardo Porotti, Antoine Essig, Benjamin Huard, and Florian Marquardt. Deep Reinforcement Learning for Quantum State Preparation with Weak Nonlinear Mea- surements. Quantum, 6:747, 2022

    2022

  24. [24]

    Bergmann, H

    K. Bergmann, H. Theuer, and B. W. Shore. Coherent population transfer among quantum states of atoms and molecules. Rev. Mod. Phys. , 70:1003–1025, 1998

    1998

  25. [25]

    Laser-induced population transfer by adiabatic passage techniques

    Nikolay Vitanov, Thomas Halfmann, Bruce Shore, and Klaas Bergmann. Laser-induced population transfer by adiabatic passage techniques. Annual review of physical chemistry, 52:763–809, 2001

    2001

  26. [26]

    Col- loquium: Coherently controlled adiabatic passage

    Petr Kr´ al, Ioannis Thanopulos, and Moshe Shapiro. Col- loquium: Coherently controlled adiabatic passage. Rev. Mod. Phys., 79:53–77, 2007

    2007

  27. [27]

    Shortcuts to adiabaticity: theoreti- cal framework, relations between different methods, and versatile approximations

    Takuya Hatomura. Shortcuts to adiabaticity: theoreti- cal framework, relations between different methods, and versatile approximations. Journal of Physics B: Atomic, Molecular and Optical Physics , 57(10):102001, 2024

    2024

  28. [28]

    Lacour, N

    X. Lacour, N. Sangouard, S. Gu´ erin, and H. R. Jauslin. Arbitrary state controlled-unitary gate by adiabatic pas- sage. Phys. Rev. A , 73:042321, 2006

    2006

  29. [29]

    I. I. Beterov, M. Saffman, E. A. Yakshina, V. P. Zhukov, D. B. Tretyakov, V. M. Entin, I. I. Ryabtsev, C. W. Mansell, C. MacCormick, S. Bergamini, and M. P. Fe- doruk. Quantum gates in mesoscopic atomic ensembles based on adiabatic passage and rydberg blockade. Phys. Rev. A, 88:010303, 2013

    2013

  30. [30]

    Quantum computation by adiabatic evo- lution

    Edward Farhi, Jeffrey Goldstone, Sam Gutmann, and Michael Sipser. Quantum computation by adiabatic evo- lution. arXiv preprint quant-ph/0001106 , 2000

    Pith/arXiv arXiv 2000

  31. [31]

    Quantum annealing in the transverse ising model

    Tadashi Kadowaki and Hidetoshi Nishimori. Quantum annealing in the transverse ising model. Physical Review E, 58(5):5355, 1998

    1998

  32. [32]

    A fast quantum mechanical algorithm for database search

    Lov K Grover. A fast quantum mechanical algorithm for database search. In Proceedings of the twenty-eighth annual ACM symposium on Theory of computing , pages 212–219, 1996

    1996

  33. [33]

    Gu´ ery-Odelin, A

    D. Gu´ ery-Odelin, A. Ruschhaupt, A. Kiely, E. Tor- rontegui, S. Mart´ ınez-Garaot, and J. G. Muga. Shortcuts to adiabaticity: Concepts, methods, and applications. Rev. Mod. Phys. , 91:045001, 2019

    2019

  34. [34]

    Dynamical invariants for quantum control of four-level systems

    Utkan G¨ ung¨ ord¨ u, Yidun Wan, Mohammad Ali Fasihi, and Mikio Nakahara. Dynamical invariants for quantum control of four-level systems. Phys. Rev. A , 86:062312, 2012

    2012

  35. [35]

    An exact quantum theory of the time-dependent harmonic oscillator and of a charged particle in a time-dependent electromagnetic field

    H Ralph Lewis Jr and WB Riesenfeld. An exact quantum theory of the time-dependent harmonic oscillator and of a charged particle in a time-dependent electromagnetic field. Journal of mathematical physics , 10(8):1458–1473, 1969

    1969

  36. [36]

    Fast optimal frictionless atom cooling in harmonic traps: Short- cut to adiabaticity

    Xi Chen, A Ruschhaupt, Sebastian Schmidt, Adolfo del Campo, David Gu´ ery-Odelin, and J Gonzalo Muga. Fast optimal frictionless atom cooling in harmonic traps: Short- cut to adiabaticity. Physical review letters , 104(6):063002, 2010

    2010

  37. [37]

    Fast atomic transport without vibrational heating

    E Torrontegui, S Ib´ a˜ nez, Xi Chen, A Ruschhaupt, D Gu´ ery-Odelin, and JG Muga. Fast atomic transport without vibrational heating. Physical Review AAtomic, Molecular, and Optical Physics , 83(1):013415, 2011

    2011

  38. [38]

    Engineering of fast population transfer in three-level systems

    Xi Chen and JG Muga. Engineering of fast population transfer in three-level systems. Physical Review AAtomic, Molecular, and Optical Physics , 86(3):033405, 2012

    2012

  39. [39]

    Lewis-riesenfeld invariants and transitionless quantum driving

    Xi Chen, Erik Torrontegui, and Juan Gonzalo Muga. Lewis-riesenfeld invariants and transitionless quantum driving. Physical Review AAtomic, Molecular, and Optical Physics, 83(6):062116, 2011

    2011

  40. [40]

    Sarandy, E.I

    M.S. Sarandy, E.I. Duzzioni, and R.M. Serra. Quantum computation in continuous time using dynamic invariants. Physics Letters A , 375(38):3343–3347, 2011

    2011

  41. [41]

    Nonadiabatic geometric quantum computation with cat- state qubits via invariant-based reverse engineering

    Yi-Hao Kang, Ye-Hong Chen, Xin Wang, Jie Song, Yan Xia, Adam Miranowicz, Shi-Biao Zheng, and Franco Nori. Nonadiabatic geometric quantum computation with cat- state qubits via invariant-based reverse engineering. Phys. Rev. Res., 4:013233, 2022

    2022

  42. [42]

    Sarandy, Eduardo I

    Marcela Herrera, Marcelo S. Sarandy, Eduardo I. Duzzioni, and Roberto M. Serra. Nonadiabatic quan- tum state engineering driven by fast quench dynamics. Phys. Rev. A , 89:022323, 2014

    2014

  43. [43]

    Shaken not stirred: creating exotic angular momentum states by shaking an optical lattice

    Anthony Kiely, Albert Benseny, Thomas Busch, and An- dreas Ruschhaupt. Shaken not stirred: creating exotic angular momentum states by shaking an optical lattice. Journal of Physics B: Atomic, Molecular and Optical Physics, 49(21):215003, 2016

    2016

  44. [44]

    Mart´ ınez-Garaot, E

    S. Mart´ ınez-Garaot, E. Torrontegui, Xi Chen, and J. G. Muga. Shortcuts to adiabaticity in three-level systems using lie transforms. Phys. Rev. A , 89:053408, May 2014

    2014

  45. [45]

    Torrontegui, S

    E. Torrontegui, S. Mart´ ınez-Garaot, and J. G. Muga. Hamiltonian engineering via invariants and dynamical algebra. Phys. Rev. A , 89:043408, 2014

    2014

  46. [46]

    Noise resistant quan- tum control using dynamical invariants

    Amikam Levy, Anthony Kiely, Juan Gonzalo Muga, Ron- nie Kosloff, and Erik Torrontegui. Noise resistant quan- tum control using dynamical invariants. New Journal of Physics, 20(2):025006, 2018

    2018

  47. [47]

    Control of open quantum systems via dynamical invariants

    Loris Maria Cangemi, Hilario Espin´ os, Ricardo Puebla, Erik Torrontegui, and Amikam Levy. Control of open quantum systems via dynamical invariants. arXiv preprint arXiv:2311.13164, 2024

    Pith/arXiv arXiv 2024

  48. [48]

    Goswami, C

    K. Goswami, C. Giarmatzi, C. Monterola, S. Shrapnel, J. Romero, and F. Costa. Experimental characterization of a non-markovian quantum process. Phys. Rev. A , 104:022432, 2021. 18

    2021

  49. [49]

    Singularities in Physics and Engineering (Second Edition)

    Paramasivam Senthilkumaran. Singularities in Physics and Engineering (Second Edition) . 2053-2563. IOP Pub- lishing, 2024

    2053

  50. [50]

    Man- aging singular behavior in the tracking control of quantum dynamical observables

    Wusheng Zhu, Martina Smit, and Herschel Rabitz. Man- aging singular behavior in the tracking control of quantum dynamical observables. The Journal of Chemical Physics , 110(4):19051915, January 1999

    1999

  51. [51]

    Singularity-free quantum tracking control of molecular rotor orientation

    Alicia Magann, Tak-San Ho, and Herschel Rabitz. Singularity-free quantum tracking control of molecular rotor orientation. Physical Review A , 98(4), October 2018

    2018

  52. [52]

    Dynamical invariant formalism of shortcuts to adiabaticity

    Kazutaka Takahashi. Dynamical invariant formalism of shortcuts to adiabaticity. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engi- neering Sciences, 380(2239), November 2022

    2022

  53. [53]

    Li, G.-C

    C.-F. Li, G.-C. Guo, and J. Piilo. Non-markovian quan- tum dynamics: What does it mean? Europhysics Letters, 127(5):50001, 2019

    2019

  54. [54]

    White, C

    G. White, C. Hill, F. Pollock, L. Hollenberg, and K. Modi. Demonstration of non-markovian process characterisation and control on a quantum processor. Nature Communi- cations, 11, 12 2020

    2020

  55. [55]

    Generalized cumulant expansion method

    Ryogo Kubo. Generalized cumulant expansion method. Journal of the Physical Society of Japan , 17(7):1100–1120, 1962

    1962

  56. [56]

    On the generators of quantum dynamical semigroups

    Goran Lindblad. On the generators of quantum dynamical semigroups. Communications in mathematical physics , 48:119–130, 1976

    1976

  57. [57]

    Completely positive dynami- cal semigroups of n-level systems.Journal of Mathematical Physics, 17(5):821–825, 1976

    Vittorio Gorini, Andrzej Kossakowski, and Ennackal Chandy George Sudarshan. Completely positive dynami- cal semigroups of n-level systems.Journal of Mathematical Physics, 17(5):821–825, 1976

    1976

  58. [58]

    On the theory of relaxation processes

    Alfred G Redfield. On the theory of relaxation processes. IBM Journal of Research and Development , 1(1):19–31, 1957

    1957

  59. [59]

    On quantum theory of transport phe- nomena: Steady diffusion

    Sadao Nakajima. On quantum theory of transport phe- nomena: Steady diffusion. Progress of Theoretical Physics, 20(6):948–959, 1958

    1958

  60. [60]

    Ensemble method in the theory of irre- versibility

    Robert Zwanzig. Ensemble method in the theory of irre- versibility. The Journal of Chemical Physics , 33(5):1338– 1341, 1960

    1960

  61. [61]

    In situ characterization of qubit control lines: A qubit as a vector network analyzer

    Markus Jerger, Anatoly Kulikov, Z´ enon Vasselin, and Arkady Fedorov. In situ characterization of qubit control lines: A qubit as a vector network analyzer. Phys. Rev. Lett., 123:150501, Oct 2019

    2019

  62. [62]

    Quantum control in the presence of strongly coupled non-markovian noise

    Arinta Auza, Akram Youssry, Gerardo Paz-Silva, and Alberto Peruzzo. Quantum control in the presence of strongly coupled non-markovian noise. arXiv preprint arXiv:2404.19251, 2024

    Pith/arXiv arXiv 2024

  63. [63]

    Paz-Silva and Lorenza Viola

    Gerardo A. Paz-Silva and Lorenza Viola. General transfer- function approach to noise filtering in open-loop quantum control. Phys. Rev. Lett. , 113:250501, Dec 2014

    2014

  64. [64]

    Efficient non- markovian quantum dynamics using time-evolving matrix product operators

    Aidan Strathearn, Peter Kirton, Dainius Kilda, Jonathan Keeling, and Brendon William Lovett. Efficient non- markovian quantum dynamics using time-evolving matrix product operators. Nature communications, 9(1):3322, 2018

    2018

  65. [65]

    Pollock, C´ esar Rodr´ ıguez-Rosario, Thomas Frauenheim, Mauro Paternostro, and Kavan Modi

    Felix A. Pollock, C´ esar Rodr´ ıguez-Rosario, Thomas Frauenheim, Mauro Paternostro, and Kavan Modi. Non- markovian quantum processes: Complete framework and efficient characterization. Phys. Rev. A , 97:012127, Jan 2018

    2018

  66. [66]

    Paz-Silva, Leigh M

    Gerardo A. Paz-Silva, Leigh M. Norris, and Lorenza Viola. Multiqubit spectroscopy of gaussian quantum noise. Phys. Rev. A, 95:022121, 2017

    2017

  67. [67]

    Characterization and control of open quantum systems beyond quantum noise spectroscopy

    Akram Youssry, Gerardo Paz Silva, and Chris Ferrie. Characterization and control of open quantum systems beyond quantum noise spectroscopy. npj Quantum Infor- mation, 6, 12 2020

    2020

  68. [68]

    Quantum feature space of a qubit coupled to an arbitrary bath

    Chris Wise, Akram Youssry, Alberto Peruzzo, Jo Plested, and Matt Woolley. Quantum feature space of a qubit coupled to an arbitrary bath. 2025

    2025

  69. [69]

    Paz-Silva, and Alberto Peruzzo

    Yule Mayevsky, Akram Youssry, Ritik Sareen, Gerardo A. Paz-Silva, and Alberto Peruzzo. Quantum engineering of qudits with interpretable machine learning. 2025

    2025

  70. [70]

    Modeling and control of a reconfigurable photonic circuit using deep learning

    Akram Youssry, Robert Chapman, Alberto Peruzzo, Chris Ferrie, and Marco Tomamichel. Modeling and control of a reconfigurable photonic circuit using deep learning. Quantum Science and Technology , 5:025001, 01 2020

    2020

  71. [71]

    The Bloch sphere plots were constructed using Quantum Toolbox in Python (QuTiP) [73]

    and Keras [ 72] Python packages. The Bloch sphere plots were constructed using Quantum Toolbox in Python (QuTiP) [73]. Invariant Design As presented in Section IV A, we design the subtrajectories based on the target state. Par- ticularly, target (i) and (ii) do not require subdividing the evolution trajectory. Targets (iii), (iv) and (v) require two subtr...

  72. [72]

    Experimental graybox quantum system identi- fication and control

    Akram Youssry, Yang Yang, Robert J Chapman, Ben Haylock, Francesco Lenzini, Mirko Lobino, and Alberto Peruzzo. Experimental graybox quantum system identi- fication and control. npj Quantum Information , 10(1):9, 2024

    2024

  73. [73]

    Opti- mal control of quantum thermal machines using machine learning

    Ilia Khait, Juan Carrasquilla, and Dvira Segal. Opti- mal control of quantum thermal machines using machine learning. Phys. Rev. Res. , 4:L012029, Mar 2022

    2022

  74. [74]

    Y.X. Zeng, J. Shen, S.C. Hou, T. Gebremariam, and C. Li. Quantum control based on machine learning in an open quantum system. Physics Letters A , 384(35):126886, 2020

    2020

  75. [75]

    Multi- axis control of a qubit in the presence of unknown non- markovian quantum noise

    Akram Youssry and Hendra Ishwara Nurdin. Multi- axis control of a qubit in the presence of unknown non- markovian quantum noise. Quantum Science and Tech- nology, 8, 2022

    2022

  76. [76]

    Noise detection with spectator qubits and quantum feature engineering

    Akram Youssry, Gerardo A Paz-Silva, and Christopher Ferrie. Noise detection with spectator qubits and quantum feature engineering. New Journal of Physics , 25(7):073004, 2023

    2023

  77. [77]

    Lutchyn, Cody P

    /suppress Lukasz Cywi´ nski, Roman M. Lutchyn, Cody P. Nave, and S. Das Sarma. How to enhance dephasing time in superconducting qubits. Phys. Rev. B , 77:174509, May 2008

    2008

  78. [78]

    Mart´ ın Abadi, Ashish Agarwal, Paul Barham, Eugene Brevdo, Zhifeng Chen, Craig Citro, Greg S. Corrado, Andy Davis, Jeffrey Dean, Matthieu Devin, Sanjay Ghe- mawat, Ian Goodfellow, Andrew Harp, Geoffrey Irving, Michael Isard, Yangqing Jia, Rafal Jozefowicz, Lukasz Kaiser, Manjunath Kudlur, Josh Levenberg, Dandelion Man´ e, Rajat Monga, Sherry Moore, Derek M...

    2015

  79. [79]

    Fran¸ cois Chollet et al. Keras. https://keras.io, 2015

    2015

  80. [80]

    Neill Lambert, Eric Gigure, Paul Menczel, Boxi Li, Patrick Hopf, Gerardo Surez, Marc Gali, Jake Lishman, Rushi- raj Gadhvi, Rochisha Agarwal, Asier Galicia, Nathan Shammah, Paul Nation, J. R. Johansson, Shahnawaz Ahmed, Simon Cross, Alexander Pitchford, and Franco Nori. Qutip 5: The quantum toolbox in python. 2024

    2024

Showing first 80 references.